Abstract Major depressive disorder (MDD) is one of the most common psychiatric disorders and a leading cause of disability. MDD is estimated to affect about 16 million Americans. Although most MDD patients respond to treatments (psychotherapy/medication), a significant subset (~25-30%) does not and is referred to as having ?treatment-resistant major depression? (TRMD). TRMD places a disproportionately heavy burden on total MDD costs from suicide, personal financial costs, loss of work, and medical comorbidity; hence, developing new treatments to target TRMD is consistent with the National Institute of Mental Health?s (NIMH) mission of reducing the burden of mental illness through research. Most existing antidepressant medications target (via neurotransmitter receptor) sites in the brain identified many years ago. Due to high prevalence and lack of adequate response in TRMD, new treatments which target novel brain regions/neurotransmitter systems are imperative if we are to successfully treat TRMD. One promising new class of TRMD treatments targets the N-methyl-D-aspartate glutamate receptor (NMDAR). In the past 10 years, many TRMD patients have responded to the NMDAR antagonist ketamine. Ketamine studies are ongoing; however, it has several limitations including cognition problems, high abuse/ addiction potential, and possible psychotic induction. Recently, our group demonstrated that another NMDAR antagonist nitrous oxide (?laughing gas?) is an effective and well-tolerated rapidly acting antidepressant in TRMD. This potentially significant finding is, however, currently lacking mechanistic understanding. Before conducting a large scale clinical efficacy trial of a new agent/treatment, NIMH recommends that scientists first determine how the agent produces an effect; that is, it is necessary to define a ?biological target? of the treatment. This current proposal uses state-of-the-art brain imaging to take the first step in understanding how nitrous oxide modulates brain function. Using functional connectivity magnetic resonance imaging (fcMRI), recent studies have demonstrated that certain networks/regions of the human brain increase their communications (i.e., ?connectivity?) in MDD. Studies also demonstrate that antidepressant treatments attenuate this increased connectivity. This study proposes to use fcMRI to study immediate (2 hours) and sustained (24 hours) effects of nitrous oxide on specific brain networks/regions (the default mode, affective, and cognitive control networks and dorsal pre-frontal nexus, amygdala, and hippocampus) identified as critical in MDD. Nondepressed and TRMD subjects will undergo fcMRI scans immediately before, and 2 and 24 hours after one hour inhalations of either nitrous oxide or a placebo. We will compare how these two exposures influence these networks/regions immediately (2 hours; Aim 1), and 24 hours after nitrous inhalation (Aim 2), and how these changes differ between cohorts. Study findings will inform future studies exploring how nitrous oxide affects the TRMD brain.